---

In Public Health, Social Science and Demography, large-scale surveys often follow a hierarchical structure of data as the surveys are based on multistage stratified cluster sampling. The appropriate way to analyzing such survey data is therefore based on nested sources of variability which come from different levels of hierarchy. Many researchers do not take into account these nested sources of variability for such multistage stratified cluster sampling dataset. Using the 2004 Bangladesh Demographic and Health Survey contraceptive binary data this work is designed to assist in all aspects of working with multilevel logistic regression models, including model conceptualization, model description, understanding of the structure of required multilevel data, estimation of the model via the statistical package MLwiN, and interpretation of the results. The used MLwiN and SPSS code have been given in the appendix. The data analysis of this work also ensures helping other analysts who work with other softwares and programmings. The analysis should be useful to professionals, MSc or PhD students working especially in the fields of Social Sciences, Demography, Public Health and Statistics.

---

He is a lecturer in Applied Statistics, ISRT, University of Dhaka, Bangladesh. Currently he is doing PhD in Statistics at Warwick University, UK. His research mainly focuses on the areas of Bayesian Statistics, Biostatistics and Social Statistics. He is an applied statistician. He is also interested in the area of Data Analysis.

---